Access to spatial information is essential for many daily activities, yet it remains a significant
challenge for blind and visually impaired people (BVIP). This paper investigates user interface (UI)
concepts for conveying spatial information and explores current research trends, emerging
technologies, and regional differences in the field. A Systematic Literature Review (SLR) was
conducted, combined with Latent Dirichlet Allocation (LDA) for topic modeling to identify key research
areas and categorize existing approaches. Findings suggest an evolving research focus driven by
advancements in mobile applications, sensor integration, augmented and virtual reality (AR/VR), and
machine learning. Regional differences and shifting trends highlight the need for innovative solutions
that address affordability, accessibility, and interoperability. This paper aims to provide a
comprehensive overview of the current state of research while identifying gaps and future
opportunities in the development of spatial user interfaces for BVIP.

In the Netherlands, approximately 451,900 people suffer from stroke effects, with 80% experiencing reduced arm-hand dexterity [1]. These individuals benefit from a rehabilitation tool that can be used independently, encouraging frequent use and providing direct progress insights, such as increased hand or finger strength. HandForce (Haptic Force Feedback on Object-manipulation in Real and Computational Environments) aims to design a hand rehabilitation tool with the following objectives: 1) home use by clients, 2) visual feedback on a computer screen or tablet, 3) Force feedback to clients and 4) integration with Virtual Reality (VR) games when possible, Although technologies like VR, gaming, and motion tracking are largely available, they are not yet specifically applicable to hand rehabilitation. Key elements such as touch, grip force, frictional resistance, and object resistance are minimally represented in the virtual world. This project explores how hand-object manipulation can be simulated in a virtual environment to match the real, physical experience, making it a useful tool for hand rehabilitation. Outcome measures and user requirements are assessed with hand therapists and patients.

The project consortium is formed by 2 knowledge institutions, 9 SMEs, and 2 rehab centres.

Psychological outcomes are known to influence participation in leisure-time physical activity (LTPA) for people with disabilities (PWD). Virtual Reality (VR) may help target important psychological outcomes for PWD, but there has been limited documentation to date. The aim of this study is to identify outcomes used to assess psychological factors and to summarize the influence of VR among PWD. A scoping review was done and found six articles examining psychological outcomes, reporting positive impacts of VR on presence, motivation and arousal. Future research is needed to further explore how VR may enhance motivation, a major predictor of LTPA behavior.

The aim of this scoping review is to provide an overview of the types of AI-enabled Orientation &

Mobility Technologies (AIOMTs) that have been developed for use by blind and visually impaired

individuals and explore the multifaceted role of trust in its adoption and use. By synthesizing existing

research, the review identifies the functional roles played by AIOMTs and key factors influencing

trust, including system design, user interaction, and contextual considerations. Furthermore, this

work examines how technology design, functional role, and trust impact the adoption, efficacy,

and long-term engagement with AIOMTs. The findings provide a comprehensive framework for

future research and development, emphasizing user-centered approaches and trust calibration

mechanisms.

This research seeks to help blind individuals by developing an AI powered assistive technology to minimize common distraction issues in guide dogs. Using a machine learning model, we can now accurately detect when a guide dog gets distracted, by processing real-time camera footage. Three recent advances have made this research possible: (1) the development of X-Y coordinate pose estimation for animals via DeepLabCut, (2) Long Short-Term Memory (LSTM) models that generalize patterns within data, and (3) GPU computational power available on demand through Google Colab. Our research methodology focuses on analyzing hours of guide dog footage, extracting and mapping body part coordinates across frames, and training an ML model to recognize postural indicators of distraction events. The model, which can then be applied to real-time camera footage, through a chest mounted device, to predict, with high probability, when a distraction event is likely to occur, providing an early warning system for visually impaired guide dog handlers.

Mobility is the process of moving form a source location to destination in safe and reliable manner. It is a precursor for seeking education and employment. Consider an example of walking to school in a city, where a person may need to take turns, follow the road/sidewalk, avoid incoming persons/obstacles finally reach the entry point of the destination. Because of the absence of visual sense, it is a day-to-day problem for blind and visually impaired (BVI) people. First we identify four key aspects of route navigation namely (i) high-level mobility, (ii) low-level mobility, (iii) path correction, and (iv) last-mile navigation and collate the challenges faced by visually impaired persons. We conduct a survey of (i) literature, (ii) products, as well as small (but representational) set of user interaction to assess the use and challenges with the assistive technologies available for each aspect. Our findings reveal that majority of research efforts have focused on obstacle avoidance and high-level navigation separately. The problem of reliably homing to a landmark separating the outdoor and indoor navigation remains largely unaddressed. These findings can help in designing navigation aids that aim to provide holistic solution to assist BVI pedestrians in outdoor environment.